Parents play a central role in efforts to prevent obesity in early childhood. Despite what is known about how stress affects obesity and about how parenting shapes children's stress response, no studies have examined how the emotional quality of observed parent-child interaction and children's attachment security are related to obesity. The epidemiologic research proposed in this application utilizes data from two large national studies, the Early Childhood Longitudinal Study, Birth Cohort (ECLS-B) and the NICHD Study of Early Child Care and Youth Development (SECCYD). The objective of the research is to determine how the emotional quality of early parent-child relationships affects risk for childhood obesity. The central hypothesis is that low levels of maternal sensitivity and responsiveness, children's observed negative behavior and lack of engagement with their mother, and insecure attachment, will increase children's risk for obesity before they enter elementary school, and be associated with higher weight gain into adolescence. The rationale for this research is that it will support development of new strategies to prevent obesity in early childhood. These strategies would move beyond the usual focus on diet and physical activity to target the quality of the early parent-child relationship, affecting energy balance by improving children's ability to respond to stress. Three specific aims will be pursued corresponding to three key predictors of children's weight: 1) maternal sensitivity and responsiveness, 2) child engagement and behavior, and 3) attachment security. These aspects of parenting, child behavior, and attachment were assessed in both studies at 24 months of age using validated measures based on direct observations of parent-child interaction. Children's heights and weights were also measured repeatedly in both studies. Levels of these predictors will be examined in relation to two outcomes: 1) risk of obesity at kindergarten (in ECLS-B using logistic regression), and 2) BMI trajectory through adolescence (in SECCYD using hierarchical linear mixed effects models). A component of each aim is assessment of effect modification to investigate the extent to which associations differ by sociodemographic characteristics. The project is innovative because it applies concepts established in the field of developmental psychology to the public health challenge of childhood obesity and moves the research field beyond its current examination of obesity and parenting, which is largely focused on parenting practices that are directly related to children's food intake and activity, or parenting styles that are based on self-report. This project will identify and characterize new risk factors for early childhood obesity that focus on the emotional quality of the parent-child relationship. The proposed epidemiologic research is significant because the results can be used to improve existing obesity prevention strategies for preventing obesity before children enter elementary school. These new strategies can be expected to have an important positive impact on children's weight and energy balance while also benefiting their social and emotional development.